摘要
为了更为精确地分析随机删失情况下故障数据的可靠性,本文以统计学理论为支撑,提出将期望-极大值(Exception maximization,EM)算法应用于随机删失数据的还原处理过程中。采集B737NG飞机IDG部件故障数据,利用K-M估计法及MATLAB软件确定其寿命分布类型为两参数威布尔分布;利用EM算法估计随机删失情况下威布尔分布的未知参数;进行可靠性及维修性分析。实证分析结果表明,威布尔分布下考虑删失数据时的可靠性分析精度更高,并根据该结论对B737NG飞机IDG部件维修计划的制定提出具体建议。
In order to analyze the reliability of failure data under random censored more accurately,an reduction method based Exception Maximization(EM)algorithm for random censored data are proposed,which based on the statistics.Firstly,collecting failure data of IDG component of B737NG aircraft,determine that the life distribution type is two parameter Weibull distribution with K-M estimation method and MATLAB software;secondly,estimating the unknown parameters of Weibull distribution by EM algorithm for random censored data;finally,analyzing reliability and maintainability of IDG failure data.The analysis results show that the reliability analysis accuracy is much higher in the case of considering censored data under the Weibull distribution,the new results can be used in developing maintenance plan and optimizing maintenance scheme for IDG of B737NG aircraft.
作者
孔祥芬
张利寒
蔡峻青
Kong Xiangfen;Zhang Lihan;Cai Junqing(Aeronautical Engineering Institute,Civil Aviation University of China,Tianjin 300300,China)
出处
《机械科学与技术》
CSCD
北大核心
2019年第7期1138-1142,共5页
Mechanical Science and Technology for Aerospace Engineering
基金
航空科学基金项目(20170267002)资助
关键词
随机删失
EM算法
可靠性
威布尔分布
MATLAB
random censored
EM algorithm
reliability
Weibull distribution
MATLAB
K-M estimation
